Several lines of evidence provide support for a model of abnormal connectivity to explain the underlying neuropathology of autism. However, no study has provided converging evidence of both functional and structural connectivity abnormalities in individuals with autism spectrum disorders (ASD). In this proposal, functional magnetic resonance imaging (fMRI) and diffusion tensor imaging (DTI) will be used to investigate abnormalities in neural connectivity in ASD. We will study two domains of impairment (face perception and response to social information) and determine how neural connectivity relates to symptom severity. Twenty-five individuals with high functioning ASD (FSIQ >80) and 25 typically developing (TD) individuals will participate in the study. The TD group will be matched to the ASD group according to age, gender, and FSIQ. White matter structural abnormalities (reduced fractional anisotropy) are hypothesized to be correlated to abnormal functional connectivity in ASD. The degree of abnormality in measures of structural and functional connectivity is expected to be mediated by clinical severity. A Career Development Award is requested in order to develop the skills necessary to conduct multimodal imaging research. The applicant will obtain training in DTI and develop methods for combining and statistically analyzing imaging data from different imaging modalities. Consultation and training from experts in the fields of neuroimaging and autism spectrum disorders will be augmented by coursework and seminars. The implementation of this training plan will yield a cohesive and thorough assessment of the integrity of white matter connections that underlie specific impairments in ASD.